Detecting and Mitigating the Effect of Manipulated Reputation on Online Social Networks
|
|
- Angel McCormick
- 8 years ago
- Views:
Transcription
1 Detecting and Mitigating the Effect of Manipulated Reputation on Online Social Networks Anupama Aggarwal «supervised by Dr. Ponnurangam Kumaraguru» Indraprastha Institute of Information Technology, Delhi New Delhi, India ABSTRACT In recent times, online social networks (OSNs) are being used not only to communicate but to also create a public/social image. Artists, celebrities and even common people are using social networks to build their brand value and gain more visibility either amongst a restricted set of people or public. In order to enable user to connect to other users in the OSN and gain following and appreciation from them, various OSNs provide different social metrics to the user such as Facebook likes, Twitter followers and Tumblr reblogs. Hence, these metrics give a sense of social reputation to the OSN user. As more users are trying to leverage social media to create a brand value and become more influential, spammers are luring such users to help manipulate their social reputation with the help of paid service (black markets) or collusion networks. In this work, we aim to build a robust alternate social reputation system and detect users with manipulated social reputation. In order to do so, we first start by understanding the underlying structure of various sources of crowdsourced social reputation manipulation like blackmarkets, supplydriven microtask websites and collusion networks. We then build a mechanism for an early detection of users with manipulated social reputation. Our initial results are encouraging and substantiate the possibility of a robust social reputation system. Keywords social reputation; online social networks; user behaviour; crowdsourced manipulation 1. PROBLEM The social network revolution has led to the rise of several online social networks (OSNs) like Facebook, Twitter and Yelp 1 which have rapidly acquired millions of users. OSN users use these services to communicate with other Copyright is held by the International World Wide Web Conference Committee (IW3C2). IW3C2 reserves the right to provide a hyperlink to the author s site if the Material is used in electronic media. WWW 16 Companion, April 11 15, 2016, Montréal, Québec, Canada. ACM /16/04. users, spread and consume information, and even build and strengthen social connections. Due to the growing rise of OSNs as a platform for mass communication, brands, celebrities and political parties have started using them extensively to engage with users. Some of the most popular and influential celebrities have garnered millions of users following their activities on OSNs. 2 The user following and crowdsourced ratings give an OSN user a sense of social reputation which she tries to maintain and boost to be more influential in the network and attract more following. Various OSNs have a different measure of user following and social reputation, like followers on Twitter and Instagram, likes on Facebook and ratings on Yelp. However, these reputation metrics can be manipulated in several ways. One of the most prevalent methods to alter social reputation is online blackmarket that not only generates a misleading reputation but also injects the network with false identities. There exist several online services from where an OSN user can purchase bulk followers and likes. Such services are very cheap and let users choose from different packages like 1000 Followers for $3. 3 Security researchers have recently estimated that the revenue generated by blackmarket for Twitter followers is between $40 million to $360 million. Another host for fake accounts peddlers is supplydriven microtask markets like Fiverr and SEOClerks. Account peddlers exploit these services to cater to OSN users who inflate their social media metrics such as followers, likes and shares in the hope to become more influential and popular on the network. These marketplaces often provide newly created or stockpiled, fake and inactive accounts to the users. Such services are a big threat to the credibility of the social networks that rely on crowdsourced ratings and reviews for product recommendation and build user trust. Recently Amazon sued 1,114 sellers on Fiverr for posting fake reviews of Amazon products [15]. Infiltration of fake accounts and metrics also has a damaging effect on the advertisement revenue framework of OSNs. An advertisement might be shown to certain users with seemingly high popularity and reputation, but will most likely not get clicked by the expected number of real users due to the accounts being fake or bots. To address such issues and mitigate the damaging effect on the OSN ecosystem, there is a dire need to build alternate social reputation systems and weed out users and entities on social networks with manipulated followers, likes and ratings. 2 Barack Obama has 67.2M Twitter followers, Shakira has 103.8M Facebook likes
2 the help of sybil nodes (fake or compromised) [16]. Other studies have also uncovered ways to detect various sources of crowdsourced manipulation of social reputation involving microtask supply-driven markets or cheap online labour markets [8, 12]. As first part of this work, we want to exhaustively landscape the sources of social reputation manipulation. We have landscaped blackmarket websites to understand who the sellers, customers and the sybil nodes are and explain them in more detail in section 5.1. Detect sybil nodes on online social networks. There has been significant research focusing on identification of sybil nodes in a network. Researchers have developed techniques to distinguish between sybil and legitimate identities and effectively identify inorganic behaviour on social networks using supervised, as well as unsupervised machine learning methods [2, 9, 20]. These techniques largely rely on the network and behavioral features of each sybil entity itself. In this work, instead of focusing on detection of each sybil node, we focus on the damage caused by a group of such sybil nodes and hence aim to detect a crowdsourced manipulation of social reputation done by these sybil nodes. Bimal et al. have looked at detection of crowdsourced tampering of a social media metric like Twitter follower count or Yelp ratings by sybil entities by finding an anomalous pattern in the joining date of all the OSN nodes involved in computing the like of rating of each entity under suspicion [19]. While this methodology proves to be effective to some extent, it needs details of each node participating in contributing to the social reputation score of the suspicious user. As the second part in this work, we want to build an effective mechanism to detect OSN users with manipulated reputation score. Our preliminary results based on a heuristic approach and only the attributes of the suspicious user seem promising, and we describe them in more detail in section 5.3. There exist few alternate reputation systems like Klout and PeerIndex, but they are not robust and yield poor results. The next step based on these results would be to build an alternate social reputation system for each OSN, which is robust and is not affected by crowdsourced manipulation of number of followers, likes or shares. Such an alternate social reputation system can be helpful to mitigate the adverse effects on OSN s ecosystem. Effect of crowdsourced manipulation of social reputation. There have been several reports stating the adverse effects of manipulated social reputation on popular OSNs like generating fake following by politicians [5], misleading product reviews [15] and fake hits on Facebook advertisements [4]. Researchers have shown that blackmarkets for crowdsourced social reputation on OSNs have a potential to generate revenue between $40 to $360 million [14]. However, the effect of social reputation manipulation of users on OSN s friend recommender system, prioritization of search results and advertising revenue of OSN is largely unexplored and not quantified. As the final part of this work, we would measure the adverse effect of manipulated social reputation on the OSN s ecosystem. Figure 1: Illustration of three step study objectives to mitigate manipulated social reputation on online social networks Recent studies show that there exist a thriving blackmarket to create fake identities [12, 21]. There have also been strong evidence of browser malwares that attack social networking sites and trigger activities such as liking and sharing posts without user s knowledge [10, 11]. The attackers behind blackmarket or browser malwares either create fake identities or compromise existing users to manipulate social reputation in bulk. These sybil identities are used to manipulate social reputation, such as Twitter followers [17], Facebook likes [18] and Yelp reviews and ratings [13]. Although there is a strong evidence of manipulated social reputation on various networks and its damaging effects [19], there does not exist alternate social reputation systems to compute the difference between percieved and real reputation of an OSN user. Coupled with this gap, there exist a lack of understanding of how manipulated social reputation affects OSN s ecosystem. An artificially inflated social reputation can adversely affect OSN s recommender system, post engagement and advertising revenue framework, which we plan to study. Therefore, my dissertation explores the following Building a robust alternate Social Reputation System which does not get affected by manipulated social metrics of the OSN user. To address the above, this work is broadly divided into three parts, also illustrated in Figure 1 Landscape the sources of manipulation of social reputation on OSNs - online blackmarkets, collusion networks, browser malwares Detect and measure the effect of sybil nodes in OSNs contributing to the fake social reputation; and an early detection mechanism for users with suspicious social reputation Use meta-information of an entity on OSN to draw out differences between perceived and real social reputation and build an alternate social reputation system for OSNs which is robust 2. STATE OF THE ART Identifying sources of social reputation manipulation. Researchers have lately discovered several sources of social reputation manipulation on OSNs which adversely affect the economy and tarnish the network s credibility. Stringhini et al. detected several thousand blackmarket websites which either create collusion networks of participating users or boost Twitter follower count of paying user with 3. BACKGROUND & APPROACH Every online social network provides mechanisms to its users to become socially visible and influential. This work explores the sources of crowdsourced manipulation of social reputation and detection of the same. Table 1 gives a brief overview of popular OSNs and correspondingly, which met- 294
3 ric gives the concerned OSN entity a sense of social reputation. The larger goal of this work is to be able to effectively detect and measure the effects of manipulation of such social reputation metrics. Table 1: Metrics on online social networks which give a sense of social reputation OSN OSN Entity Perceived Reputation Score Twitter, Instagram, Pinterest User Follower Count Facebook Page Likes Yelp, Foursquare Business/ Venue Rating Amazon, Flipkart, Ebay Product Rating Tumblr Post Reblogs YouTube Video Likes To achieve our goal as described in section 3, we divide our work into a three phases. We describe our current and future approach to complete each phase in more detail as followed 3.1 Landscaping the Source of Social Reputation Manipulation We identify three main sources of bulk social reputation manipulation viz. (i) Online Blackmarket, (ii) Scratch-Back Services, (iii) Browser Malwares. There has been some work done to identify Online Blackmarkets by finding websites on search engines using keywords like buy cheap followers and purchase Facebook likes [16]. Online blackmarkets are however not hard to find since they advertise themselves to get more hits. We focus on the more challenging task of profiling and measuring the impact of these websites. We identify market leaders that inject the network with most of the sybil nodes and hence have a more damaging effect on the network. Another source of social reputation manipulation, Scratch- Back services are the ones where users do not need to pay money to boost their social reputation. Instead, they can do favors to other interested users to get back the same and hence become part of a collusion network. Since each individual node involved in such a collusion network can be a genuine user trying to manipulate its social reputation, such manipulation is very hard to detect. There do not exist any studies with tangible results to uncover large-scale collusion networks involved in manipulating social reputation. However, we plan to take cues from seminal research been done on social spam campaigns [6, 3] to detect social reputation manipulation by scratch-back services. Browser Malwares is one of the other primary sources of social reputation manipulation which we plan to study. The infected users are unaware of the likes, shares and comments triggered from their accounts and unknowingly become a proxy for a manipulated social reputation of the attacker. This area is largely unexplored with only evidence of browser malwares attacking OSNs [7]. We plan to collect a dataset of rogue browser extensions that trigger such attacks and study the propagation of infection in the network. 3.2 Early Detection of Users with Manipulated Social Reputation While there have been studies to detect individual sybil nodes involved in crowdsourced manipulation, there has been very less research to effectively identify OSN users with manipulated social reputation. Bimal et al. proposed to identify crowdsourced manipulation by investigating the timestamp of activities by involved sybil nodes [19]. Since it is not feasible to fetch the timestamp in almost every OSN of the past activity, they rely on the joining dates distribution. Their assumption is that the distribution of joining dates of nodes involved in a crowdsourced manipulation will have significant divergence from that of a reliable reference set of high-quality nodes. However, the disadvantage of this approach is that it will fail in case the sybil nodes are stockpiled and hence there won t be any anomaly in their joining dates distribution. We propose a robust approach to detect crowdsourced social reputation manipulation by leveraging the basics of each OSN used to gain true attention and reputation in a network. For instance, a Twitter user will genuinely garner followers when she posts interesting content (or is a celebrity in the real world). Therefore, to validate her follower count, we build a heuristic to determine her real reputation based on various other factors like tweet-follower ratio, evidence of being a topic expert and overlap of her interests with followers. The initial results of our approach seem promising and based on our findings we later plan to propose an alternate social reputation score. 3.3 Measuring the Effects of Manipulated Social Reputation on OSN s Ecosystem Measuring and quantifying the adverse effects of crowdsourced manipulation remains an unexplored area. We plan to conduct the following experiments to understand the effect on OSN s ecosystem Effect on Recommender Systems: Do more people follow a user after her social reputation is artificially boosted? Does she have a higher chance to get suggested by friend/follow recommender system of the social network? Effect on Social Capital of User: Does she become more impactful and visible? Does a manipulated social reputation also bump up her existing alternate social reputation scores like Klout and PeerIndex? Effect during Events: Does the user s posts about a trending event get higher priority in search results after her social reputation is manipulated? Does it trigger a chain reaction garnering higher engagement and attention from other users following the event? 4. EVALUATION Our three-fold study will be evaluated as followed To effectively profile the sources of crowdsourced manipulation of social reputation, we will collect a true positive large-scale dataset of online blackmarket websites, scratch-back services and browser malwares. To understand how these services operate, we plan to create dummy accounts and use these services. As initial part of our experiments, we created dummy ac- 295
4 counts and purchased Twitter followers from blackmarket websites. 4 We plan to similarly collect large scale data of scratch-back services and browser malwares for an effective evaluation To evaluate our approach to detect users with crowdsourced manipulation of social reputation, we plan to compare our technique with (i) existing social reputation systems like Klout and PeerIndex, and (ii) methodology proposed in existing work by Bimal et al. [19]. We plan to conduct in-the-wild online experiments to study the effect of crowdsourced social reputation manipulation. Collecting temporal data with multiple snapshots before and after the manipulation will help us evaluate how effectively we can measure and detect the changes in effects on OSN s ecosystem. 5. RESULTS We have following results for the experiments described in the previous sections 5.1 Landscaping the Online Blackmarket The first part involves landscaping the sources of crowdsourced social reputation manipulation. As a preliminary study, we have been able to effectively study the underlying structure of online blackmarket which sells Twitter followers. We purchased over fake followers from over 60 most popular (by Alexa ranking) merchant websites and discovered the following We defined Quality of Service (QoS) metric for blackmarket services and discovered that 95% of these services have a QoS score of 0.28 (on a scale of 0 to 1, with 1 being highest) or lesser. While this is not surprising since such services are not reliable and may not deliver as promised, they still attract heavy traffic as shown by Alexa rankings. We use several parameters like spread on social networks, promotion by affected users and Alexa rankings to determine the market leaders. We discovered that only a few merchant websites contribute towards more than 70% popularity and usage for manipulation of Twitter follower count. This shows that there is an underlying oligopolistic structure, and we found evidence of collusion among these leading merchant websites. Figure 2 shows this phenomenon in more detail. More details on landscaping the blackmarket services can be found in our recently published work [1]. 5.2 Detection of Fake Twitter Followers To understand how the sybil nodes created by the blackmarket services operate, we studied their behaviour in initial part of our work. To conduct the experiment, we used the same data purchased for the experiment in the previous section. With respect to these sybil nodes, in this case, fake (purchased) Twitter followers, we discovered the following 4 We ensured that all money we paid to underground merchants to acquire fake followers was exclusively for the dummy Twitter accounts we created. The dummy accounts were fully controlled by us and were for the sole purpose of conducting experiments. Figure 2: Popularity and spread of blackmarket websites selling fake Twitter followers. Few sellers are market leaders Fake followers have low social engagement. These users do not indulge in conversations with other users. We also found that they do not post original content and most of the tweets consist of RTs (Retweets). One of the other striking features of fake followers was that they have followers and followees tweeting in several languages. We found that 13% percent users had followees tweeting in 5 or more languages. We can detect fake following behaviour from legitimate with an accuracy of 89.2% by using temporal behavioral and network features of the suspicious users. Some of the most important attributes we discovered were number of unfollows by suspicious user over time, language overlap with the users she follows and social engagement like RTs Table 2 elaborates results of the evaluation metrics for detection of suspicious following behaviour. Table 2: Confusion Matrix Classification Results of distinguishing legitimate users from those exhibiting suspicious following behaviour. True Predicted Suspicious Legitimate Suspicious Legitimate Detailed explanation of characteristics of fake followers can be seen in our recent work [1]. 5.3 Detection of Users with Manipulated Social Reputation In this section we move towards the more challenging task of effective detection of users with manipulated social reputation. We conduct our initial experiment on Twitter. The perceived social reputation is defined by the follower count of a user. To assess the credibility of this follower count, we use the following parameters: tweet-follower ratio, evidence of user being a topic expert and overlap of her interests with followers. To build our baseline, we collected a random sample of 1.6 million Twitter users and used a multiple linear regression model to fit the three parameters. We define our model as 296
5 y i = β 0 + β 1x i1 + β 2x i2 + β 3x i3 For a particular user under suspicion, we find the deviation of its distribution based on the same three parameters from that of the random sample and define suspect ratio as the inverse of the deviation. Therefore, smaller suspect ratio indicates higher suspicion towards the perceived social reputation (follower count) of the user. We then conducted an in-the-wild experiment over 1% Twitter stream data and labeled a user as suspicious if her suspect ratio was significantly low below a certain threshold. So far we have been able to label over 56,000 users as suspicious. A small sample of the users detected by our proposed methodology can be seen at 6. CONCLUSION AND FUTURE WORK This work aims to detect and measure the deviation from perceived social reputation of an OSN user. We start by landscaping the sources of such manipulation like blackmarkets and scratch-back services. Preliminary results bring out the underlying structure of blackmarket which can be helpful to uncover the market leaders. Eliminating or hindering their operations can significantly bring down crowdsourced manipulation of social reputation. Initial results also show that a robust and adaptive technique can be built to detect social reputation manipulation. However, our proposed framework is at a very nascent stage and needs much more improvement and rigorous evaluation. Much work yet remains to leverage this framework to build an alternate social reputation system and measure the effects of social reputation manipulation on OSN s ecosystem. 7. REFERENCES [1] A. Aggarwal and P. Kumaraguru. What they do in shadows: Twitter underground follower market. In Privacy, Security and Trust (PST), [2] F. Benevenuto, G. Magno, T. Rodrigues, and V. Almeida. Detecting spammers on twitter. In Collaboration, electronic messaging, anti-abuse and spam conference (CEAS), volume 6, page 12, [3] Z. Chu, I. Widjaja, and H. Wang. Detecting social spam campaigns on twitter. In Applied Cryptography and Network Security, pages Springer, [4] D. B. Clark. The bot bubble. April [5] DailyMail. More than 2 million of hillary clinton s twitter followers are fake or never tweet /More-2-MILLION-Hillary-Clinton-s- Twitter-followers, April [6] H. Gao, J. Hu, C. Wilson, Z. Li, Y. Chen, and B. Y. Zhao. Detecting and characterizing social spam campaigns. In Proceedings of the 10th ACM SIGCOMM conference on Internet measurement, pages ACM, [7] A. Kapravelos, C. Grier, N. Chachra, C. Kruegel, G. Vigna, and V. Paxson. Hulk: Eliciting malicious behavior in browser extensions. In Proceedings of the 23rd Usenix Security Symposium, [8] K. Lee, S. Webb, and H. Ge. Characterizing and automatically detecting crowdturfing in fiverr and twitter. Social Network Analysis and Mining, 5(1):1 16, [9] E.-P. Lim, V.-A. Nguyen, N. Jindal, B. Liu, and H. W. Lauw. Detecting product review spammers using rating behaviors. In Proceedings of the 19th ACM international conference on Information and knowledge management, pages ACM, [10] Microsoft. Trojan:js/febipos.a. encyclopedia/entry.aspx?name=trojan: JS/Febipos.A, August [11] Microsoft. Trojan:js/kilim.a. threat/encyclopedia/entry.aspx?name=js/kilim, June [12] M. Motoyama, D. McCoy, K. Levchenko, S. Savage, and G. M. Voelker. Dirty jobs: The role of freelance labor in web service abuse. In Proceedings of the 20th USENIX conference on Security, pages USENIX Association, [13] NYTimes. A rave, a pan, or just a fake? May [14] NYTimes. Fake twitter followers become multimillion-dollar business. April [15] NYTimes. All the product reviews money can buy. December [16] G. Stringhini, G. Wang, M. Egele, C. Kruegel, G. Vigna, H. Zheng, and B. Y. Zhao. Follow the green: growth and dynamics in twitter follower markets. In Proceedings of the 2013 conference on Internet measurement conference, pages ACM, [17] K. Thomas, D. McCoy, C. Grier, A. Kolcz, and V. Paxson. Trafficking fraudulent accounts: The role of the underground market in twitter spam and abuse. In USENIX Security, pages Citeseer, [18] B. Viswanath, M. A. Bashir, M. Crovella, S. Guha, K. P. Gummadi, B. Krishnamurthy, and A. Mislove. Towards detecting anomalous user behavior in online social networks. In Proceedings of the 23rd USENIX Security Symposium (USENIX Security), [19] B. Viswanath, M. A. Bashir, M. B. Zafar, S. Bouget, S. Guha, K. P. Gummadi, A. Kate, and A. Mislove. Strength in numbers: Robust tamper detection in crowd computations. In Proceedings of the 2015 ACM on Conference on Online Social Networks, pages ACM, [20] G. Wang, T. Wang, H. Zheng, and B. Y. Zhao. Man vs. machine: Practical adversarial detection of malicious crowdsourcing workers. In 23rd USENIX Security Symposium, USENIX Association, CA, [21] G. Wang, C. Wilson, X. Zhao, Y. Zhu, M. Mohanlal, H. Zheng, and B. Y. Zhao. Serf and turf: crowdturfing for fun and profit. In Proceedings of the 21st international conference on World Wide Web, pages ACM,
Detection of Malicious URLs by Correlating the Chains of Redirection in an Online Social Network (Twitter)
International Journal of Research Studies in Computer Science and Engineering (IJRSCSE) Volume 1, Issue 3, July 2014, PP 33-38 ISSN 2349-4840 (Print) & ISSN 2349-4859 (Online) www.arcjournals.org Detection
More informationHow To Measure Spam On Multiple Platforms
Observing Common Spam in Tweets and Email Cristian Lumezanu NEC Laboratories America Princeton, NJ lume@nec-labs.com Nick Feamster University of Maryland College Park, MD feamster@cs.umd.edu ABSTRACT Spam
More informationAn analysis of the effectiveness of personalized spam using online social network public information
An analysis of the effectiveness of personalized spam using online social network public information Enaitz Ezpeleta 1, Urko Zurutuza 1, and José María Gómez Hidalgo 2 1 Electronics and Computing Department,
More informationFollow the Green: Growth and Dynamics in Twitter Follower Markets
Follow the Green: Growth and Dynamics in Twitter Follower Markets Gianluca Stringhini, Gang Wang, Manuel Egele, Christopher Kruegel, Giovanni Vigna, Haitao Zheng, Ben Y. Zhao Department of Computer Science,
More informationURL Shortening in Twitter
Improved URL Shortening With Traffic Analysis and Spam Filter 1 Ms. S.Rajeshwari, 2 R.Banupriya, 3 M.Manibharathi, 4 M.Samena 1 AP/CSE, 2 UG Scholar Department of Computer Science and Engineering Jay Shriram
More informationTrafficking Fraudulent Accounts
Trafficking Fraudulent Accounts The Role of the Underground Market in Twitter Spam & Abuse Kurt Thomas, Damon McCoy, Chris Grier, Alek Kolcz, Vern Paxson UCB, GMU, ICSI, Twitter Overview Google, Facebook,
More informationEnhancing Twitter Spam Accounts Discovery Using Cross-Account Pattern Mining
Enhancing Twitter Spam Accounts Discovery Using Cross-Account Pattern Mining Ioana-Alexandra Bara, Carol J. Fung, Thang Dinh Department of Computer Science, Virginia Commonwealth University, Virginia,
More informationNetwork-based spam filter on Twitter
Network-based spam filter on Twitter Ziyan Zhou Stanford University ziyanjoe@stanford.edu Lei Sun Stanford University sunlei@stanford.edu ABSTRACT Rapidly growing micro-blogging social networks, such as
More informationMimicking human fake review detection on Trustpilot
Mimicking human fake review detection on Trustpilot [DTU Compute, special course, 2015] Ulf Aslak Jensen Master student, DTU Copenhagen, Denmark Ole Winther Associate professor, DTU Copenhagen, Denmark
More informationSpam Filtering in Twitter using Sender-Receiver Relationship
Spam Filtering in Twitter using Sender-Receiver Relationship Jonghyuk Song 1, Sangho Lee 1 and Jong Kim 2 1 Dept. of CSE, POSTECH, Republic of Korea {freestar,sangho2}@postech.ac.kr 2 Div. of ITCE, POSTECH,
More informationFraud Detection in Online Reviews using Machine Learning Techniques
ISSN (e): 2250 3005 Volume, 05 Issue, 05 May 2015 International Journal of Computational Engineering Research (IJCER) Fraud Detection in Online Reviews using Machine Learning Techniques Kolli Shivagangadhar,
More informationRemoving Web Spam Links from Search Engine Results
Removing Web Spam Links from Search Engine Results Manuel EGELE pizzaman@iseclab.org, 1 Overview Search Engine Optimization and definition of web spam Motivation Approach Inferring importance of features
More informationMarti Motoyama, Damon McCoy, Kirill Levchenko, Stefan Savage, and Geoffrey M. Voelker UC San Diego 1
Marti Motoyama, Damon McCoy, Kirill Levchenko, Stefan Savage, and Geoffrey M. Voelker UC San Diego 1 Many Web services today are free/open access Supported by advertising revenue Reaching critical mass
More informationMEASURING AND FINGERPRINTING CLICK-SPAM IN AD NETWORKS
MEASURING AND FINGERPRINTING CLICK-SPAM IN AD NETWORKS Vacha Dave *, Saikat Guha and Yin Zhang * * The University of Texas at Austin Microsoft Research India Internet Advertising Today 2 Online advertising
More informationMALLET-Privacy Preserving Influencer Mining in Social Media Networks via Hypergraph
MALLET-Privacy Preserving Influencer Mining in Social Media Networks via Hypergraph Janani K 1, Narmatha S 2 Assistant Professor, Department of Computer Science and Engineering, Sri Shakthi Institute of
More informationLASTLINE WHITEPAPER. Using Passive DNS Analysis to Automatically Detect Malicious Domains
LASTLINE WHITEPAPER Using Passive DNS Analysis to Automatically Detect Malicious Domains Abstract The domain name service (DNS) plays an important role in the operation of the Internet, providing a two-way
More informationExploring Big Data in Social Networks
Exploring Big Data in Social Networks virgilio@dcc.ufmg.br (meira@dcc.ufmg.br) INWEB National Science and Technology Institute for Web Federal University of Minas Gerais - UFMG May 2013 Some thoughts about
More informationClick Traffic Analysis of Short URL Spam on Twitter
Click Traffic Analysis of Short URL Spam on Twitter De Wang, Shamkant B. Navathe, Ling Liu, Danesh Irani, Acar Tamersoy, Calton Pu College of Computing Georgia Institute of Technology Atlanta, Georgia
More informationVictimless Malware How Blackhats Make a Killing Targeting Companies
Victimless Malware How Blackhats Make a Killing Targeting Companies Lou Manousos RiskIQ Session ID: HT2-401 Session Classification: Intermediate Serendipity Simple Browser Tricks, Hacks and Kits Large
More informationWE KNOW IT BEFORE YOU DO: PREDICTING MALICIOUS DOMAINS Wei Xu, Kyle Sanders & Yanxin Zhang Palo Alto Networks, Inc., USA
WE KNOW IT BEFORE YOU DO: PREDICTING MALICIOUS DOMAINS Wei Xu, Kyle Sanders & Yanxin Zhang Palo Alto Networks, Inc., USA Email {wei.xu, ksanders, yzhang}@ paloaltonetworks.com ABSTRACT Malicious domains
More informationDNS Changer Remediation Study
DNS Changer Remediation Study The following study was presented by Georgia Tech researchers at the M 3 AAWG 27 th General Meeting February 19, 2013, San Francisco M 3 AAWG Messaging, Malware and Mobile
More informationSpam Detection on Twitter Using Traditional Classifiers M. McCord CSE Dept Lehigh University 19 Memorial Drive West Bethlehem, PA 18015, USA
Spam Detection on Twitter Using Traditional Classifiers M. McCord CSE Dept Lehigh University 19 Memorial Drive West Bethlehem, PA 18015, USA mpm308@lehigh.edu M. Chuah CSE Dept Lehigh University 19 Memorial
More informationBig Data and Cyber Security A bibliometric study Jacky Akoka, Isabelle Comyn-Wattiau, Nabil Laoufi Workshop SCBC - 2015 (ER 2015) 1 Big Data a new generation of technologies and architectures, designed
More informationWe Know It Before You Do: Predicting Malicious Domains
We Know It Before You Do: Predicting Malicious Domains Abstract Malicious domains play an important role in many attack schemes. From distributing malware to hosting command and control (C&C) servers and
More informationLearn How to Defend Your Online Marketplace from Unwanted Traffic
Learn How to Defend Your Online Marketplace from Unwanted Traffic Speakers Rami Essaid CEO & Co-founder Distil in Classifieds The Basics of Bots A Bot is an automated program that runs on the internet
More informationRecurrent Patterns Detection Technology. White Paper
SeCure your Network Recurrent Patterns Detection Technology White Paper January, 2007 Powered by RPD Technology Network Based Protection against Email-Borne Threats Spam, Phishing and email-borne Malware
More informationInternet Monitoring via DNS Traffic Analysis. Wenke Lee Georgia Institute of Technology
Internet Monitoring via DNS Traffic Analysis Wenke Lee Georgia Institute of Technology 0 Malware Networks (Botnets) 1 From General-Purpose to Targeted Attacks 11/14/12 2 Command and Control l Botnet design:
More informationEVILSEED: A Guided Approach to Finding Malicious Web Pages
+ EVILSEED: A Guided Approach to Finding Malicious Web Pages Presented by: Alaa Hassan Supervised by: Dr. Tom Chothia + Outline Introduction Introducing EVILSEED. EVILSEED Architecture. Effectiveness of
More informationTop 10 Anti-fraud Tips: The Cybersecurity Breach Aftermath
ebook Top 10 Anti-fraud Tips: The Cybersecurity Breach Aftermath Protecting against downstream fraud attacks in the wake of large-scale security breaches. Digital companies can no longer trust static login
More informationRole of Social Networking in Marketing using Data Mining
Role of Social Networking in Marketing using Data Mining Mrs. Saroj Junghare Astt. Professor, Department of Computer Science and Application St. Aloysius College, Jabalpur, Madhya Pradesh, India Abstract:
More informationAn analysis of various tools, methods and systems to generate fake accounts for social media
An analysis of various tools, methods and systems to generate fake accounts for social media A Thesis Proposal Presented by Avanish Pathak to the faculty of College of Computer and Information Science
More informationOpen an attachment and bring down your network?
Open an attachment and bring down your network? Many people think this will never happen to them, but virus attacks can come from unlikely sources and can strike when you least expect it. They can wreak
More informationStreamlining Web and Email Security
How to Protect Your Business from Malware, Phishing, and Cybercrime The SMB Security Series Streamlining Web and Email Security sponsored by Introduction to Realtime Publishers by Don Jones, Series Editor
More informationReverse Engineering Socialbot Infiltration Strategies in Twitter
Reverse Engineering Socialbot Infiltration Strategies in Twitter Carlos Freitas, Fabricio Benevenuto, Saptarshi Ghosh and Adriano Veloso Universidade Federal de Minas Gerais, Belo Horizonte, Brazil Max
More informationEnterprise-Grade Security from the Cloud
Datasheet Website Security Enterprise-Grade Security from the Cloud Unmatched web application security experience, enhanced by real-time big data analytics, enables Incapsula to provide best-of-breed security
More informationAppWatcher: Unveiling the Underground Market of Trading Mobile App Reviews
AppWatcher: Unveiling the Underground Market of Trading Mobile App Reviews ABSTRACT Zhen Xie Dept of Computer Science and Engineering The Pennsylvania State University zhenxie@cse.psu.edu Driven by huge
More informationBusiness Process Services. White Paper. Social Media Influence: Looking Beyond Activities and Followers
Business Process Services White Paper Social Media Influence: Looking Beyond Activities and Followers About the Author Vandita Bansal Vandita Bansal is a subject matter expert in Analytics and Insights
More informationSOCIAL MEDIA FOR MSMEs A turning point. By DR. PRALAY DEY National Small Industries Corporation (NSIC)
SOCIAL MEDIA FOR MSMEs A turning point By DR. PRALAY DEY National Small Industries Corporation (NSIC) IMPORTANCE OF MSMEs in India 44+ MILLIONS UNITS 45% manufacturing output MSMEs in India 40% EXPORT
More informationSearch Engine Optimization (SEO)
Search Engine Optimization (SEO) Saurabh Chavan, Apoorva Chitre, Husain Bhala Abstract Search engine optimization is often about making small modifications to parts of your website. When viewed individually,
More informationContent-Based Discovery of Twitter Influencers
Content-Based Discovery of Twitter Influencers Chiara Francalanci, Irma Metra Department of Electronics, Information and Bioengineering Polytechnic of Milan, Italy irma.metra@mail.polimi.it chiara.francalanci@polimi.it
More informationWHITE PAPER. Social media analytics in the insurance industry
WHITE PAPER Social media analytics in the insurance industry Introduction Insurance is a high involvement product, as it is an expense. Consumers obtain information about insurance from advertisements,
More informationLASTLINE WHITEPAPER. Large-Scale Detection of Malicious Web Pages
LASTLINE WHITEPAPER Large-Scale Detection of Malicious Web Pages Abstract Malicious web pages that host drive-by-download exploits have become a popular means for compromising hosts on the Internet and,
More informationA Novel Distributed Denial of Service (DDoS) Attacks Discriminating Detection in Flash Crowds
International Journal of Research Studies in Science, Engineering and Technology Volume 1, Issue 9, December 2014, PP 139-143 ISSN 2349-4751 (Print) & ISSN 2349-476X (Online) A Novel Distributed Denial
More informationSponsorship & Social/ Digital Media Brent Barootes May 28, 2013. WHITE PAPER Presented by Partnership Group Sponsorship Specialists
Sponsorship & Social/ Digital Media Brent Barootes May 28, 2013 WHITE PAPER Presented by Partnership Group Sponsorship Specialists SPONSORSHIP AND SOCIAL / DIGITAL MEDIA HOW THESE NEW MEDIUMS CAN MAXIMIZE
More informationT: 0800 135 7186 cravenplan.com/search
Introduction Cravenplan Computers Ltd has been building and optimising websites for over 12 years and with a dedicated, experienced search engine marketing team we are in an excellent position to help
More informationSocial Influence Analysis in Social Networking Big Data: Opportunities and Challenges. Presenter: Sancheng Peng Zhaoqing University
Social Influence Analysis in Social Networking Big Data: Opportunities and Challenges Presenter: Sancheng Peng Zhaoqing University 1 2 3 4 35 46 7 Contents Introduction Relationship between SIA and BD
More informationDetecting Spam Bots in Online Social Networking Sites: A Machine Learning Approach
Detecting Spam Bots in Online Social Networking Sites: A Machine Learning Approach Alex Hai Wang College of Information Sciences and Technology, The Pennsylvania State University, Dunmore, PA 18512, USA
More informationCS 558 Internet Systems and Technologies
CS 558 Internet Systems and Technologies Dimitris Deyannis deyannis@csd.uoc.gr 881 Heat seeking Honeypots: Design and Experience Abstract Compromised Web servers are used to perform many malicious activities.
More informationIntroduction The Case Study Technical Background The Underground Economy The Economic Model Discussion
Internet Security Seminar 2013 Introduction The Case Study Technical Background The Underground Economy The Economic Model Discussion An overview of the paper In-depth analysis of fake Antivirus companies
More information10 Things Every Web Application Firewall Should Provide Share this ebook
The Future of Web Security 10 Things Every Web Application Firewall Should Provide Contents THE FUTURE OF WEB SECURITY EBOOK SECTION 1: The Future of Web Security SECTION 2: Why Traditional Network Security
More informationTHE DIGITAL DIRECTION: HOTEL MARKETING BUDGETS AND A DIGITAL 101 FOR HOTELS PART I OF A TWO- PART SERIES
273 Walt Whitman Road #321, Huntington Station, NY 11746 tel: 516.680.8529 Leora@LHLcommunications.com LHLcommunications.com FEBRUARY 2014 THE DIGITAL DIRECTION: HOTEL MARKETING BUDGETS AND A DIGITAL 101
More informationMalicious Websites uncover vulnerabilities (browser, plugins, webapp, server), initiate attack steal sensitive information, install malware, compromise victim s machine Malicious Websites uncover vulnerabilities
More informationthe beginner s guide to SOCIAL MEDIA METRICS
the beginner s guide to SOCIAL MEDIA METRICS INTRO Social media can be an incredibly important business tool. Tracking the right social metrics around your industry, company, products, competition and
More informationEffective Software Security Management
Effective Software Security Management choosing the right drivers for applying application security Author: Dharmesh M Mehta dharmeshmm@mastek.com / dharmeshmm@owasp.org Table of Contents Abstract... 1
More informationDesign and Evalua.on of a Real- Time URL Spam Filtering Service
Design and Evalua.on of a Real- Time URL Spam Filtering Service Kurt Thomas, Chris Grier, Jus.n Ma, Vern Paxson, Dawn Song University of California, Berkeley Interna.onal Computer Science Ins.tute Mo.va.on
More informationGFI White Paper. How Web Reputation increases your online protection
GFI White Paper How Web Reputation increases your online protection Contents Introduction to Web Reputation 3 Why use Web Reputation? 3 The value of using Web Reputation and antivirus software 3 The value
More informationQUARTERLY REPORT 2015 INFOBLOX DNS THREAT INDEX POWERED BY
QUARTERLY REPORT 2015 INFOBLOX DNS THREAT INDEX POWERED BY EXPLOIT KITS UP 75 PERCENT The Infoblox DNS Threat Index, powered by IID, stood at 122 in the third quarter of 2015, with exploit kits up 75 percent
More informationMARKETING KUNG FU SEO: Key Things to Expand Your Digital Footprint. A Practical Checklist
MARKETING KUNG FU SEO: Key Things to Expand Your Digital Footprint A Practical Checklist 1 1. Content Development... Page 3 2. Content Organization... Page 4 3. META Data... Page 5 4. Fix Errors... Page
More informationECM 210 Chapter 6 - E-commerce Marketing Concepts: Social, Mobile, Local
Consumers Online: The Internet Audience and Consumer Behavior Around 75% (89 million) of U.S. households have Internet access in 2012 Intensity and scope of use both increasing Some demographic groups
More informationCommtouch RPD Technology. Network Based Protection Against Email-Borne Threats
Network Based Protection Against Email-Borne Threats Fighting Spam, Phishing and Malware Spam, phishing and email-borne malware such as viruses and worms are most often released in large quantities in
More informationWEB ATTACKS AND COUNTERMEASURES
WEB ATTACKS AND COUNTERMEASURES February 2008 The Government of the Hong Kong Special Administrative Region The contents of this document remain the property of, and may not be reproduced in whole or in
More informationTop Cyber Threats Of 2009
Top Cyber Threats Of 2009 Who were the top 5 riskiest celebrities in 2009? Did spammers really know who killed Michael Jackson? Data from Symantec Security Response November 2009 Top 5 Riskiest Celebrities
More informationSocial marketing and sustainability
Social marketing and sustainability There are no limits to the possibilities now March 6, 2014 Learning Objectives and Agenda Learning Objectives: 1. Educate leadership about the value or potential of
More informationJoint Industry Committee for Web Standards JICWEBS. Social Video Code of Conduct
Joint Industry Committee for Web Standards JICWEBS Social Video Code of Conduct Issued February 2015 THE SOCIAL VIDEO CODE OF CONDUCT CONTENTS The Social Video Code of Conduct comprises: Page 1 Compliance
More informationDigital Marketing Proposal. www.digitalgateway.in
Digital Marketing Proposal What is digital marketing? Digital marketing is the process of building and maintaining customer relationships through online activities to facilitate the exchange of ideas,
More informationsmart. uncommon. ideas.
smart. uncommon. ideas. Executive Overview Your brand needs friends with benefits. Content plus keywords equals more traffic. It s a widely recognized onsite formula to effectively boost your website s
More informationApplication of Machine Learning and Crowdsourcing. to Detection of Cybersecurity Threats
Application of Machine Learning and Crowdsourcing to Detection of Cybersecurity Threats February 2011 Eugene Fink, Mehrbod Sharifi, and Jaime G. Carbonell eugenefink@cmu.edu, mehrbod@cs.cmu.edu, jgc@cs.cmu.edu
More informationFinding Hidden Gems in the App Ecosystem
Whitepaper Finding Hidden Gems in the App Ecosystem How appbackr and Appthority Sort Through Millions of Apps to Deliver a Curated List of the World s Best Apps August 2013 Introduction The mobile app
More information5 Point Social Media Action Plan.
5 Point Social Media Action Plan. Workshop delivered by Ian Gibbins, IG Media Marketing Ltd (ian@igmediamarketing.com, tel: 01733 241537) On behalf of the Chambers Communications Sector Introduction: There
More informationGeneral Guidance Note on Digital Marketing and Promotions
General Guidance Note on Digital Marketing and Promotions Who should read this? Any provider that promotes its premium rate service (PRS) using forms of digital marketing. What is the purpose of this Guidance?
More informationPresented by Spyros Mygdanis & Giorgos Kalaitzis
Presented by Spyros Mygdanis & Giorgos Kalaitzis About this presentation 1. Websites Elements of a successful website 2. Content 3. Internet Marketing Channels 4. Building your online strategy 1. Websites
More informationEXPOSING CLICK FRAUD
Using Web Analytics To Identify Possible Click Fraud A White Paper From Net Applications NETAPPLICATIONS.COM What Is Click Fraud? page 2/7 Click Fraud Vs. Paid Placement "In your experience, how much of
More informationHow to Identify Phishing E-Mails
How to Identify Phishing E-Mails How to recognize fraudulent emails and avoid being phished. Presented by : Miguel Fra, Falcon IT Services (miguel@falconitservices.com) http://www.falconitservices.com
More informationRecommended Practice Case Study: Cross-Site Scripting. February 2007
Recommended Practice Case Study: Cross-Site Scripting February 2007 iii ACKNOWLEDGEMENT This document was developed for the U.S. Department of Homeland Security to provide guidance for control system cyber
More informationHTTPS Inspection with Cisco CWS
White Paper HTTPS Inspection with Cisco CWS What is HTTPS? Hyper Text Transfer Protocol Secure (HTTPS) is a secure version of the Hyper Text Transfer Protocol (HTTP). It is a combination of HTTP and a
More informationan Essential Marketing Grow Your Business Online: The Commercial Approach to Search Engine Marketing Prepared by Flex4, December 2011, v1.
A Science, a Black Art or an Essential Marketing tool? Grow Your Business Online: The Commercial Approach to Search Engine Marketing Prepared by Flex4, December 2011, v1. Grow Your Business Online P a
More informationCAPTCHA-based DDoS Defense System of Call Centers against Zombie Smart-Phone
CAPTCHA-based DDoS Defense System of Call Centers against Zombie Smart-Phone Seung Wook Jung Soongsil University seungwookj@ssu.ac.kr Abstract Recently, some researchers alarm the security community that
More informationThe 6 Step Customer Advocate Marketing Handbook
The 6 Step Customer Advocate Marketing Handbook How to Motivate and Scale Advocates to Grow Your Business I THIS BRAND How to Motivate and Scale Advocates to Grow Your Business Intro: Part 1: Why Powering
More informationModern Cyber Threats. how yesterday s mind set gets in the way of securing tomorrow s critical infrastructure. Axel Wirth
Modern Cyber Threats how yesterday s mind set gets in the way of securing tomorrow s critical infrastructure Axel Wirth Healthcare Solutions Architect Distinguished Systems Engineer AAMI 2013 Conference
More informationRecognizing Spam. IT Computer Technical Support Newsletter
IT Computer Technical Support Newsletter March 23, 2015 Vol.1, No.22 Recognizing Spam Spam messages are messages that are unwanted. If you have received an e-mail from the Internal Revenue Service or the
More information#umea2014 Digital Strategy
#umea2014 Digital Strategy Digital Personality Tone of Umeå 2014 Personality The communication should feel alive, inspiring, playful and current. Social media is one of the forces helping the culture to
More informationDetecting Social Spam Campaigns on Twitter
Detecting Social Spam Campaigns on Twitter Zi Chu 1, Indra Widjaja 2, and Haining Wang 1 1 Department of Computer Science, The College of William and Mary, Williamsburg, VA 23187, USA {zichu,hnw}@cs.wm.edu
More informationWebsite Security. End-to-End Application Security from the Cloud. Cloud-Based, Big Data Security Approach. Datasheet: What You Get. Why Incapsula?
Datasheet: Website Security End-to-End Application Security from the Cloud Unmatched web application security experience, enhanced by real-time big data analytics, enables Incapsula to provide best-ofbreed
More informationInternet Marketing Institute Delhi Mobile No.: 9643815724 DIMI. Internet Marketing Institute Delhi (DIMI)
Internet Marketing Institute Delhi () LEARN INTERNET MARKETING To earn money at home and work with big entrepreneurs Batches Weekday Batches (60 hours) Morning Batches (60 hours) Evening Batches (60 hours)
More informationB2B Social Media Marketing LeadFormix Best Practices
Introduction Social media marketing is quickly becoming one of the most popular marketing techniques for B2B enterprises. Forrester Research predicts that companies will spend some $3.1 billion annually
More informationATTORNEYS AT LAW ONLINE MARKETING MAGIC INSTALLMENT 2: THE COVER-UP IS WORSE THAN THE CRIME
SILVERMAN ADVISORS ATTORNEYS AT LAW CLIENT ALERT AUGUST 2014 ONLINE MARKETING MAGIC INSTALLMENT 2: THE COVER-UP IS WORSE THAN THE CRIME BY SCOTT A. SILVERMAN Are you still impressed when you hear that
More informationTwitter Games: How Successful Spammers Pick Targets
Twitter Games: How Successful Spammers Pick Targets Vasumathi Sridharan, Vaibhav Shankar, Minaxi Gupta School of Informatics and Computing, Indiana University {vsridhar, vshankar, minaxi}@cs.indiana.edu
More informationWHITE PAPER Online Marketing Fail #32. Failing to Properly Optimize Social Media. www.apogeeresults.com. Social Media: A Tangled Web
Failing to Properly Optimize Social Media With over $2 billion per year being spent on social media marketing, it s obvious that social has become a material (and growing) component of the overall budget.
More informationA shift in influence has occurred
Influence at Scale Thanks to social media, a shift in influence has taken place evolving from a small group of endorsing celebrities, to a few thousand influential individuals with high Klout scores, to
More informationES ET DE LA VIE PRIVÉE E 29 th INTERNATIONAL CONFERENCE OF DATA PROTECTION AND PRIVACY COMMISSIONERS
ES ET DE LA VIE PRIVÉE E 29 th INTERNATIONAL CONFERENCE OF DATA PROTECTION AND PRIVACY COMMISS The Internet Threat Landscape Symantec TM Dean Turner Director Global Intelligence Network Symantec Security
More information9You can help build your customer base and attract more visitors to your ebay
TOP SEO STRATEGIES 9You can help build your customer base and attract more visitors to your ebay shop by employing search engine optimisation (SEO) tactics. Here are a few tried and tested SEO strategies
More informationTweetAttacks Pro. User Manual
TweetAttacks Pro User Manual The ultimate twitter auto follower, auto unfollower, tweet scraper, reply generator, auto retweeter, tweet spinner, mass retweeter and tweet scheduler with socialoomph integration.
More informationDetecting misinformation in online social networks using cognitive psychology
Kumar and Geethakumari Human-centric Computing and Information Sciences 2014, 4:14 RESEARCH Open Access Detecting misinformation in online social networks using cognitive psychology KP Krishna Kumar *
More informationTHE 2014 THREAT DETECTION CHECKLIST. Six ways to tell a criminal from a customer.
THE 2014 THREAT DETECTION CHECKLIST Six ways to tell a criminal from a customer. Telling criminals from customers online isn t getting any easier. Attackers target the entire online user lifecycle from
More informationMeasuring User Credibility in Social Media
Measuring User Credibility in Social Media Mohammad-Ali Abbasi and Huan Liu Computer Science and Engineering, Arizona State University Ali.abbasi@asu.edu,Huan.liu@asu.edu Abstract. People increasingly
More informationMalware, Phishing, and Cybercrime Dangerous Threats Facing the SMB State of Cybercrime
How to Protect Your Business from Malware, Phishing, and Cybercrime The SMB Security Series Malware, Phishing, and Cybercrime Dangerous Threats Facing the SMB State of Cybercrime sponsored by Introduction
More informationCredit Card Fraud Detection Using Self Organised Map
International Journal of Information & Computation Technology. ISSN 0974-2239 Volume 4, Number 13 (2014), pp. 1343-1348 International Research Publications House http://www. irphouse.com Credit Card Fraud
More informationSOCIAL LISTENING AND KPI MEASUREMENT Key Tips for Brands to Drive Their Social Media Performance
SOCIAL LISTENING AND KPI MEASUREMENT Key Tips for Brands to Drive Their Social Media Performance With social media marketing, the power is derived from being able to directly communicate with consumers
More informationA few legal comments on spamdexing
A few legal comments on spamdexing by Gerrit Vandendriessche 1 Altius, Brussels What is spamdexing? The Internet contains a lot of information. In 2002, the total number of web pages was estimated at 2.024
More informationTODAY S USER GENERATED CONTENT
TODAY S USER GENERATED CONTENT How publishers, marketers, and agencies are leveraging user generated content (UGC) in marketing campaigns as part of their content marketing mix. October, 2014 IAB Content
More information